Image resolution is one of the important indexes that evaluate image quality. In order to improve the image resolution, software processing may be performed on multiple low-resolution images obtained by photographing the same scene whose contents are similar but whose spatio and/or temporal information is not completely the same, to generate a super-resolution image. The super-resolution image is widely used for, for example, restoring high frequency information lost in different acquisition scenes, such as out-of-focus, motion blur, non-ideal sampling, etc., and can also be configured to restore high frequency spatial information beyond a diffraction limit of an optical system.
With the on-going development of imaging technologies, technologies of acquiring super-resolution images based on a camera array cause researchers to pay high attention. The camera array may also be called array-type camera or array camera, which, relative to the traditional single camera, has a wider vision and generally comprises a plurality of cameras array-distributed, wherein each camera has a separate optical lens and an image sensor. When image collection is performed on a to-be-shot scene by using a camera array, each camera in the camera array can photograph an image of the scene respectively, images obtained through photographing of different cameras have a certain vision offset therebetween, in this way, multiple images having similar contents but having scene vision can be obtained, a super-resolution image can be acquired according to the multiple images, resolution of the super-resolution image is greater than that of a single image acquired by each camera, for example, for a camera array formed by N×N cameras, suppose that the number of pixel points of an image sensor of each camera is D×D, in an ideal situation, resolution that can be generated according to the images acquired by the cameras of the camera array is ND×ND.